Breeding for better clover-Rhizobium symbiosis: selection enriches for significant biomass GWAS SNP alleles

Abstract

White clover (Trifolium repens L.) is an integral component of mixed pastures in temperate agriculture, providing quality feed and a sustainable source of fixed nitrogen (N) through its symbiosis with soil-dwelling Rhizobium bacteria. While there has been much focus on identifying and applying Rhizobium inoculants with improved N-fixation, there has been less attention on identifying and exploiting plant genetic factors to develop cultivars that routinely form effective Rhizobium symbioses. The complexity of the clover-Rhizobium interaction makes for a challenging but valuable breeding target and an important strategy for reducing N-fertiliser use.

We investigated deployment of genomic selection (GS) for clover-Rhizobium symbiotic response by growing seedlings from 120 half-sibling (HS) families in vermiculite/McKnight’s solution where N was provided either by a commercial Rhizobium strain (TA1) or supplied mineral N (positive control). Shoot dry matter (DM) among other traits was recorded after 35 days of growth. Symbiotic potential (SP) for each HS family was calculated as DM produced in the Rhizobium treatment as a proportion of the positive control DM. The traits showed significant additive variation and narrow-sense heritabilities ranged from 0.24-0.33. A Smith-Hazel (SH) multi-trait selection index to generate the largest plants with the greatest SP was derived. The maternal parents of the 120 HS families were genotyped with 110,000 genotyping by sequencing (GBS)-derived SNPs and their phenotypes inferred from performance of their respective HS family. These data underpinned prediction modelling using KGD-GBLUP, and cross-validation of the model generated predictive abilities ranging from 0.23-0.36 depending on the trait.

Crosses were made for the SH index (combined Shoot DM and SP) using i) standard Among Family selection based on family phenotype (ApHSF); and ii) Among and Within Family selection (ApWgsHSF) incorporating GS. For the GS, 40 seedlings each of the top and bottom 12 performing HS families were genotyped (GBS) and genomic estimated breeding values were calculated using KGD-GBLUP to identify predicted best individuals within families. The predicted top two individuals from each of the top 12 families were polycrossed. A divergent selection was made using the two predicted bottom performing individuals from each of the bottom 12 families. Phenotyping (see above) showed progeny of the top ApHSF cross had a mean shoot dry matter 58% greater than progeny from the bottom ApHSF cross. Incorporating GS increased this difference to 95% between progenies of the top and bottom ApWgsHSF crosses.

To investigate associated markers and underlying trait mechanisms, a genome-wide association study (GWAS) of the 120 families identified a significant SNP each on Chromosomes 2 and 16 with four other SNPs approaching the Bonferroni threshold. The most significant SNP was near a pentatricopeptide repeat gene associated with regulating reactive oxygen species that influence Rhizobium symbiosis specificity. Furthermore, the GWAS SNP variants associated with increased biomass were absent or at very low frequencies in the bottom performing families. These SNPs will be assessed as markers to improve prediction models.

These data highlight that clover genetics can be harnessed for increased DM in symbiosis with Rhizobia, and the next steps include field assessment.